The Autism Genetic Resource Exchange, or AGRE, is celebrating its 10 year anniversary! Since 1997, AGRE has collected genetic and clinical information from individuals with autism and their families. AGRE now includes data from over 1600 families, and is in use by over 160 researchers from 13 countries. It has contributed significantly to the productivity of autism researchers, and AGRE reached a milestone recently when the 100th research paper citing the AGRE database was published. With a clever idea and a rich database like AGRE, researchers are identifying critical pieces of the autism puzzle. In celebration of its 10th birthday, some of the key findings using the AGRE database are highlighted here.
Tracking down the genes related to autism is challenging because of the wide variety of autism phenotypes: no two people with autism are exactly alike. AGRE helps researchers tackle the complexity of this disorder by providing a rich database that's large enough to allow the critical comparisons necessary to find mutations shared by some autistic subjects, but not by non-autistic subjects. The genes that seem related to autism (“candidate genes”) increase our understanding of what mechanisms may have gone awry in autism and can potentially be used as biomarkers for early diagnosis and biological interventions.
To find candidate genes, some researchers start with a known gene whose function seems to explain some autistic behavior, then ask whether autistic subjects have altered forms of the gene. Using this strategy, scientists at Rutgers University found an association between the EN2 gene and autism by examining mice lacking this gene because they have a having a similar brain abnormality as one seen in autism, namely a loss of the Purkinje cell types in the cerebellum. These mice lacked the EN2 gene. Another study conducted at Vanderbilt University focused on the MET gene, which contributes to brain development, immune system function, and gastrointestinal repair, three disparate realms in which symptoms co-occur in some cases of autism. Using the AGRE database, disrupted forms of these genes were found in people with autism. In the case of the MET gene, the variant associated with autism was more prevalent in families with multiple cases of autism (“multiplex”) than those with a no previous history of autism (“simplex”).
Another strategy for finding candidate genes is to ask how people with autism differ genetically from those without autism. This requires scans of thousands of genes using DNA microarray technology (“gene chips”), which can quickly determine which genes are turned on or off (“expression”). The Autism Genome Project, a collaborative research team made up of 137 researchers from over 50 institutions, looked for candidate genes within the largest collection of autism families (1,496), including those from AGRE . This scan highlighted an area on chromosome 11 and also the neurexin 1 gene, which is important for forming the connections between neurons that enable them to communicate. They also reported that gains and losses of chunks of DNA called copy number variations (CNVs) were also linked to some cases of autism. Another study from George Washington University took a more targeted approach by searching for genetic differences among the rare cases of identical twins that do not share autism . This study found that many genes involved in neural development and function were expressed differently in the twins and that their expression correlated with autism severity. Researchers from UCLA took a different approach by looking for genetic similarities among autistic subjects with Fragile X, duplication of chromosome 15, and those with idiopathic autism or autism without a known cause. This study revealed a subset of genes that were expressed similarly across all three groups. These candidate genes included those involved in communication between neurons, adding to the evidence of disrupted neural connectivity in autism.
Beyond knowing which genes are impaired in autism, it is important to find out how a person with autism acquired the defective genes in the first place. Were they inherited from parents, or did the genetic changes happen spontaneously in an egg or sperm cell (“germline”) prior to conception, perhaps influenced by environmental factors? A significant role for both inherited and spontaneous modes of autism transmission has been identified using the AGRE resource.
Researchers from Washington University in St. Louis found that the social impairments found in autism may be inherited: the non-autistic siblings of autistic individuals displayed mild social impairments that, while not of clinical concern, were not found in controls . This suggests that this autistic trait may be inherited from parents, but expressed to different degrees among the children. Meanwhile, two other studies supported a role for spontaneously occurring mutations, which are those present in autistic individuals, but not in their parents. Researchers from Cold Spring Harbor Laboratory found that new, spontaneous CNV mutations were associated with autism . The particular mutations varied for each subject, supporting the idea that autism can result from a combination of genetic mutations. What causes these spontaneous mutations is unknown, but environmental factors capable of damaging the genes inside egg or sperm cells could be involved. This study also highlighted a difference between simplex and multiplex cases of autism, finding that the new CNVs were more frequent in simplex cases than in multiplex cases. This suggested that simplex cases of autism could result from new, spontaneous mutations in the germline, whereas multiplex cases could result from inherited mutations. Consistent with this notion of germline effects, another study by researchers from UCLA and AGRE confirmed a link between advanced paternal age and autism risk , suggesting that spontaneous mutations in sperm cells from older men could increase the likelihood of autism in their offspring.
Ultimately, the wide variety of genetic and clinical data available through AGRE provided researchers with the large sample sizes and the wide diversity of clinical data on families with autism that were crucial for each particular study. With continued family participation, AGRE will continue to be a moving force in autism research. The AGRE families deserve our thanks for their participation and commitment to finding the answers to autism.
1. Benayed et al (2005) Support for the homeobox transcription factor gene ENGRAILED 2 as an autism spectrum disorder susceptibility locus. Am J Hum Genet 77(5): 851-868.
2. Campbell et al (2006) A genetic variant that disrupts MET transcription is associated with autism. Proc Nat Acad Sci USA 103(45): 16834-16839.
3. The Autism Genome Project Consortium (2007) Mapping autism risk loci using genetic linkage and chromosomal rearrangements. Nat Genet 39(3): 319-328.
4. Hu et al (2006) Gene expression profiling of lymphoblastoid cell lines from monozygotic twins discordant in severity of autism reveals differential regulation of neurologically relevant genes. BMC Genomics 7: 118.
5. Nishimura et al (2007) Genome-wide expression profiling of lymphoblastoid cell lines distinguishes different forms of autism and reveals shared pathways. Human Molecular Genetics 16(14): 1682-1698.
6. Constantino et al (2006) Autistic social impairment in the siblings of children with pervasive developmental disorders. Am J Psychiatry 163: 294-296.
7. Sebat et al (2007) Strong association of de novo copy number mutations with autism. Science 316: 445-449.
8. Cantor et al (2007) Paternal age and autism are associated in a family-based sample. Molecular Psychiatry 12: 419-423.